t0028: fix test for UTF-16-LE-BOM

According to its name, the test is designed for UTF-16-LE-BOM.
However, possibly due to copy&paste oversight, it was using UTF-32.

While the test succeeds (extra \000\000 are interpreted as NUL),
I myself had an unrelated problem which caused the test to fail.
When analyzing the failure I was quite puzzled by the fact that the
test is obviously buggy.  And it seems that I'm not alone:
https://public-inbox.org/git/CAH8yC8kSakS807d4jc_BtcUJOrcVT4No37AXSz=jePxhw-o9Dg@mail.gmail.com/T/#u

Fix the test to follow its original intention.

Signed-off-by: Alexandr Miloslavskiy <alexandr.miloslavskiy@syntevo.com>
Reviewed-by: Torsten Bögershausen <tboegi@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>

t/t0028-working-tree-encoding.sh

@@ -40,7 +40,7 @@ test_expect_success 'setup test files' '
 	printf "$text" | write_utf16 >test.utf16.raw &&
 	printf "$text" | write_utf32 >test.utf32.raw &&
 	printf "\377\376"                         >test.utf16lebom.raw &&
-	printf "$text" | iconv -f UTF-8 -t UTF-32LE >>test.utf16lebom.raw &&
+	printf "$text" | iconv -f UTF-8 -t UTF-16LE >>test.utf16lebom.raw &&
 
 	# Line ending tests
 	printf "one\ntwo\nthree\n" >lf.utf8.raw &&